Add data param to mmh3_x86_128 __init__

CHANGELOG.md

@@ -25,12 +25,15 @@ This project has adhered to
 - Change the format of CHANGELOG.md to conform to the
   [Keep a Changelog](https://keepachangelog.com/en/1.1.0/) standard
   ([#63](https://github.com/hajimes/mmh3/issues/63)).
+- **Backward-incompatible**: Change the constructors of hasher classes to
+  accept a buffer as the first argument.
 
 ### Fixed
 
 - Fix a reference leak in the `hash_from_buffer()` function
   ([#75](https://github.com/hajimes/mmh3/issues/75)).
-- Fix type hints.
+- Fix type hints ([#76](https://github.com/hajimes/mmh3/issues/76),
+  [#77](https://github.com/hajimes/mmh3/issues/77)).
 
 ## [4.1.0] - 2024-01-09
 

README.md

@@ -150,12 +150,15 @@ complete changelog.
 - Change the format of CHANGELOG.md to conform to the
   [Keep a Changelog](https://keepachangelog.com/en/1.1.0/) standard
   ([#63](https://github.com/hajimes/mmh3/issues/63)).
+- **Backward-incompatible**: Change the constructors of hasher classes to
+  accept a buffer as the first argument.
 
 #### Fixed
 
 - Fix a reference leak in the `hash_from_buffer()` function
   ([#75](https://github.com/hajimes/mmh3/issues/75)).
-- Fix type hints.
+- Fix type hints ([#76](https://github.com/hajimes/mmh3/issues/76),
+  [#77](https://github.com/hajimes/mmh3/issues/77)).
 
 ### [4.1.0] - 2024-01-09
 
@@ -166,20 +169,11 @@ complete changelog.
 #### Fixed
 
 - Fix issues with Bazel by changing the directory structure of the project
-  (<https://github.com/hajimes/mmh3/issues/50>).
-- Fix incorrect type hints (<https://github.com/hajimes/mmh3/issues/51>).
+  ([#50](https://github.com/hajimes/mmh3/issues/50)).
+- Fix incorrect type hints ([#51](https://github.com/hajimes/mmh3/issues/51)).
 - Fix invalid results on s390x when the arg `x64arch` of `hash64` or
-  `hash_bytes` is set to `False` (<https://github.com/hajimes/mmh3/issues/52>).
-
-### [4.0.1] - 2023-07-14
-
-#### Changed
-
-- Refactor the project structure (<https://github.com/hajimes/mmh3/issues/48>).
-
-#### Fixed
-
-- Fix incorrect type hints.
+  `hash_bytes` is set to `False`
+  ([#52](https://github.com/hajimes/mmh3/issues/52)).
 
 ## License
 

docs/CONTRIBUTING.md

@@ -102,6 +102,14 @@ by Python.
 by Simon Willison is a good introduction to Docker/QEMU settings for emulating
 s390x.
 
+If the above does not work, you may also want to try the following:
+
+```shell
+docker run --rm --privileged tonistiigi/binfmt --install all
+docker buildx create --name mybuilder --use
+docker run -it multiarch/ubuntu-core:s390x-focal /bin/bash
+```
+
 ## Pull request
 
 Once you've pushed your changes to your fork, you can
@@ -160,7 +168,7 @@ For example,
 pip install ".[benchmark]"
 mkdir results
 python benchmark/benchmark.py -o results/mmh3_128.json \
-            --test-hash mmh3_128 --test-buffer-size-max 134217728
+            --test-hash mmh3_128 --test-buffer-size-max 262144
 ```
 
 As of version 4.2.0, the following hash function identifiers are available for

src/mmh3/mmh3module.c

@@ -1623,62 +1623,18 @@ typedef struct {
 
 static PyTypeObject MMH3Hasher128x86Type;
 
-static void
-MMH3Hasher128x86_dealloc(MMH3Hasher128x86 *self)
-{
-    Py_TYPE(self)->tp_free((PyObject *)self);
-}
-
-static PyObject *
-MMH3Hasher128x86_new(PyTypeObject *type, PyObject *args, PyObject *kwds)
-{
-    MMH3Hasher128x86 *self;
-    self = (MMH3Hasher128x86 *)type->tp_alloc(type, 0);
-    if (self != NULL) {
-        self->h1 = 0;
-        self->h2 = 0;
-        self->h3 = 0;
-        self->h4 = 0;
-        self->buffer1 = 0;
-        self->buffer2 = 0;
-        self->buffer3 = 0;
-        self->buffer4 = 0;
-        self->shift = 0;
-        self->length = 0;
-    }
-    return (PyObject *)self;
-}
-
-static int
-MMH3Hasher128x86_init(MMH3Hasher128x86 *self, PyObject *args, PyObject *kwds)
-{
-    static char *kwlist[] = {"seed", NULL};
-
-    if (!PyArg_ParseTupleAndKeywords(args, kwds, "|I", kwlist, &self->h1))
-        return -1;
-
-    self->h2 = self->h1;
-    self->h3 = self->h1;
-    self->h4 = self->h1;
-
-    return 0;
-}
-
-static PyObject *
-MMH3Hasher128x86_update(MMH3Hasher128x86 *self, PyObject *obj)
+static FORCE_INLINE void
+update_x86_128_impl(MMH3Hasher128x86 *self, Py_buffer *buf)
 {
     Py_ssize_t i = 0;
-    Py_buffer buf;
     uint32_t h1 = self->h1;
     uint32_t h2 = self->h2;
     uint32_t h3 = self->h3;
     uint32_t h4 = self->h4;
     uint32_t k1 = 0;
 
-    GET_BUFFER_VIEW_OR_ERROUT(obj, &buf);
-
-    for (; i < buf.len; i++) {
-        k1 = ((uint8_t *)buf.buf)[i];
+    for (; i < buf->len; i++) {
+        k1 = ((uint8_t *)buf->buf)[i];
         if (self->shift < 32) {  // TODO: use bit ops
             self->buffer1 |= k1 << self->shift;
         }
@@ -1720,12 +1676,71 @@ MMH3Hasher128x86_update(MMH3Hasher128x86 *self, PyObject *obj)
         }
     }
 
-    PyBuffer_Release(&buf);
+    PyBuffer_Release(buf);
 
     self->h1 = h1;
     self->h2 = h2;
     self->h3 = h3;
     self->h4 = h4;
+}
+
+static void
+MMH3Hasher128x86_dealloc(MMH3Hasher128x86 *self)
+{
+    Py_TYPE(self)->tp_free((PyObject *)self);
+}
+
+static PyObject *
+MMH3Hasher128x86_new(PyTypeObject *type, PyObject *args, PyObject *kwds)
+{
+    MMH3Hasher128x86 *self;
+    self = (MMH3Hasher128x86 *)type->tp_alloc(type, 0);
+    if (self != NULL) {
+        self->h1 = 0;
+        self->h2 = 0;
+        self->h3 = 0;
+        self->h4 = 0;
+        self->buffer1 = 0;
+        self->buffer2 = 0;
+        self->buffer3 = 0;
+        self->buffer4 = 0;
+        self->shift = 0;
+        self->length = 0;
+    }
+    return (PyObject *)self;
+}
+
+static int
+MMH3Hasher128x86_init(MMH3Hasher128x86 *self, PyObject *args, PyObject *kwds)
+{
+    Py_buffer target_buf = {0};
+    static char *kwlist[] = {"data", "seed", NULL};
+
+    if (!PyArg_ParseTupleAndKeywords(args, kwds, "|y*I", kwlist, &target_buf,
+                                     &self->h1))
+        return -1;
+
+    self->h2 = self->h1;
+    self->h3 = self->h1;
+    self->h4 = self->h1;
+
+    if (target_buf.buf != NULL) {
+        // target_buf will be released in update_x86_128_impl
+        update_x86_128_impl(self, &target_buf);
+    }
+
+    return 0;
+}
+
+static PyObject *
+MMH3Hasher128x86_update(MMH3Hasher128x86 *self, PyObject *obj)
+{
+    Py_buffer buf;
+
+    GET_BUFFER_VIEW_OR_ERROUT(obj, &buf);
+
+    // buf will be released in update_x86_128_impl
+    update_x86_128_impl(self, &buf);
 
     Py_RETURN_NONE;
 }

tests/test_mmh3_hasher.py

@@ -20,6 +20,11 @@ def test_mmh3_32_digest() -> None:
     hasher.update(b" world!")
     assert hasher.digest() == b"\xBA\x4C\x88\x24"
 
+    hasher = mmh3.mmh3_32(b"", 0x9747B28C)
+    hasher.update(b"Hello,")
+    hasher.update(b" world!")
+    assert hasher.digest() == b"\xBA\x4C\x88\x24"
+
     hasher = mmh3.mmh3_32(b"Hello,", 0x9747B28C)
     hasher.update(b" world!")
     assert hasher.digest() == b"\xBA\x4C\x88\x24"
@@ -146,6 +151,10 @@ def test_mmh3_x64_128_digest() -> None:
     hasher.update(b"The quick brown fox jumps over the lazy dog")
     assert hasher.digest() == b"!1c\xd2;\x7f\x8as\xe5\x16\xc0~rsE\xf9"
 
+    hasher = mmh3.mmh3_x64_128(b"", 0x9747B28C)
+    hasher.update(b"The quick brown fox jumps over the lazy dog")
+    assert hasher.digest() == b"!1c\xd2;\x7f\x8as\xe5\x16\xc0~rsE\xf9"
+
     hasher = mmh3.mmh3_x64_128(b"The quick brown ", seed=0x9747B28C)
     hasher.update(b"fox jumps over the lazy dog")
     assert hasher.digest() == b"!1c\xd2;\x7f\x8as\xe5\x16\xc0~rsE\xf9"
@@ -327,6 +336,25 @@ def test_mmh3_x86_128_digest() -> None:
     hasher.update(b"zy dog")
     assert hasher.digest() == b"^\xd5\xd4\x8aqa\xb8L\x9c:\xa7\x8e>y\xb6\xcd"
 
+    hasher = mmh3.mmh3_x86_128(b"", 0x9747B28C)
+    hasher.update(b"The quick brown fox ju")
+    hasher.update(b"mps ove")
+    hasher.update(b"r the la")
+    hasher.update(b"zy dog")
+    assert hasher.digest() == b"^\xd5\xd4\x8aqa\xb8L\x9c:\xa7\x8e>y\xb6\xcd"
+
+    hasher = mmh3.mmh3_x86_128(b"The quick brown fox ju", seed=0x9747B28C)
+    hasher.update(b"mps ove")
+    hasher.update(b"r the la")
+    hasher.update(b"zy dog")
+    assert hasher.digest() == b"^\xd5\xd4\x8aqa\xb8L\x9c:\xa7\x8e>y\xb6\xcd"
+
+    hasher = mmh3.mmh3_x86_128(b"The quick brown fox ju", 0x9747B28C)
+    hasher.update(b"mps ove")
+    hasher.update(b"r the la")
+    hasher.update(b"zy dog")
+    assert hasher.digest() == b"^\xd5\xd4\x8aqa\xb8L\x9c:\xa7\x8e>y\xb6\xcd"
+
 
 def test_mmh3_x86_128_sintdigest() -> None:
     hasher = mmh3.mmh3_x86_128(seed=0x9747B28C)